J. Phys. Chem. B 2006, 110, 12119-12124
12119
Hydrophobic and Electrostatic Interactions in the Adsorption of Fibronectin at Maleic Acid Copolymer Films Toshihisa Osaki,† Lars Renner,† Manuela Herklotz,† and Carsten Werner*,†,‡ Max Bergmann Center of Biomaterials Dresden, Leibniz Institute of Polymer Research Dresden, Hohe Str. 6, 01069 Dresden, Germany, and Department of Mechanical and Industrial Engineering, UniVersity of Toronto, 5 King's College Road, Toronto, Ontario, Canada, M5S 3G8 ReceiVed: February 17, 2006
Adsorption and desorption of fibronectin (FN) were investigated at thin films of alternating maleic acid copolymers with octadecene (POMA) and with propene (PPMA). The hydrophobicity and charge density of the polymers were modulated by the choice of the comonomer. In consequence, the dominant forces between the substrate and the protein were specified as hydrophobic interaction for POMA and electrostatic interaction for PPMA. The adsorption kinetics were investigated in situ as variations of the optical thickness, adsorbed mass, and viscoelastic properties (detected by reflectometric interference spectroscopy and quartz crystal microbalance technique, respectively) while alterations of the electrosurface properties were derived from surface conductivity data and isoelectric points (by streaming potential/current measurements using a microslit electrokinetic setup). The results demonstrate that the interfacial mode of adsorbed FN depends on the predominant interactions: large amounts of FN were tightly bound to POMA by hydrophobic interactions. In contrast, FN adsorbed on PPMA was concluded to attain an unfolded structure allowing for the “electrostatic matching” of positively charged residues on FN with the maleic acid groups. This conclusion was supported by the acidic IEP of 3.2 found for FN on PPMA and a significant reduction of the surface conductivity of the FN-covered polymer film, whereas FN on POMA showed an IEP of 4.2 (close to the intrinsic IEP of FN), indicating a stochastic orientation of the adsorbed protein.
Introduction Fibronectin (FN), a large dimeric glycoprotein of around 440 kDa, is a key component of the extracellular matrix, containing binding sites for heparin, collagen, integrins, and fibrins and related to a variety of cellular states and fate decisions, for instance, cell attachment, cell spreading, matrix assembly, and wound healing.1-3 Accordingly, FN is used widely in the design of biomimetic substrates and matrixes in order to stimulate cell adhesion. However, the availability of surface-bound FN for biomolecular interactions very much depends on its interfacial status, which, in turn, depends on the characteristics of the solid surface and the contact conditions applied during the immobilization. Therefore, the detailed understanding of distinct “modes” of FN at interfaces is of paramount interest in biomaterials science.4-10 Recently, the adsorption of FN was investigated on a set of alternating maleic acid copolymer films with varied physicochemical properties by means of quartz crystal microbalance measurements with dissipation monitoring (QCM-D) and confocal laser scanning microscopy (cLSM) with a fluorescent labeling technique. The adsorption, desorption, and displacement characteristics of FN showed distinct differences between poly(octadecene-alt-maleic acid) (POMA) and poly(propene-altmaleic acid) (PPMA) thin films:11,12 the authors concluded that the strong hydrophobic binding force of POMA results in distorted/compact FN conformations while the electrostatic * To whom corresponding should be addressed. E-mail:
[email protected]. Phone: +49-351-4658-531. Fax: +49-351-4658-533. † Leibniz Institute of Polymer Research Dresden. ‡ University of Toronto.
interaction merely gives weak interactions between PPMA and FN, conserving the secondary structure of the adsorbed proteins. Also, endothelial cells were grown on the same set of FN-coated copolymer thin films,13-15 revealing that the underlying substrate influenced not only adhesion but also the differentiation of this cell type through the modulation of the interfacial FN layers. Although these findings clearly show the importance of the binding strength between the adsorbed FN and the substrates, the interfacial forces involved in the adsorption of the protein and the characteristics of the resulting FN layers deserve further investigation. A recently developed experimental approach combines electrokinetics and visible light reflectometry and allows us to analyze surface charge characteristics and variations of the optical layer thickness simultaneously at solid/liquid interfaces.16 One component of the technique, the in-house developed microslit electrokinetic setup (MES), provides surface conductivity Kσ (i.e., mobile-ion concentration at the interface) as well as the electrical potential ζ at the hydrodynamic phase boundary. The MES was extended for sensing thickness variations with the in situ reflectometric interference spectroscopy (RIfS), making use of interference patterns of reflected lights at the interfaces. The combined approach has been applied successfully for in situ studies of adsorption rates against alterations of surface charge characteristics as observed in the course of protein adsorption onto polymer films.17,18 In this study, we consider the interplay of hydrophobic and electrostatic interactions in the adsorption and desorption of FN at two different types of alternating maleic acid copolymer thin films. For the chosen POMA and PPMA films, the attractive force toward the FN was considered to be mainly either
10.1021/jp061022w CCC: $33.50 © 2006 American Chemical Society Published on Web 05/26/2006
12120 J. Phys. Chem. B, Vol. 110, No. 24, 2006
Osaki et al.
|
|
dIs/d p L KB ) h + Kσ dUs/d p 2b 2
(3)
Tubing Ag/AgCl electrodes were placed at the channel inlet and outlet for the measurements. Electrolyte solutions with desired electrolyte concentration and/or pH were prepared by addition of 0.1 M KCl, HCl, and KOH solutions to degassed deionized water. Technical features of the MES and theoretical background were described in detail previously.21-23 The additional RIfS unit (Spekol 1100, Analytik Jena AG, Germany) was assembled with the MES for the extended evaluation of time-dependent assessment of variations of the optical layer thickness at the solid/liquid interface. A visible light spectrometer (halogen light source) was connected to the bottom of one of the sample carriers by an optical fiber (see Figure 1). The incident light is partly reflected at both interfaces of the targeting layer, that is, the carrier (bottom) side and the adjacent liquid (top) side, according to the individual refractive indices. The optical thickness of the layer is determined by one of the extrema of the obtained interference pattern Figure 1. Schematic representation of streaming potential/current measurements combined with reflectometric interference spectroscopy for the characterization of interfacial processes.
hydrophobic or electrostatic, respectively. The impact of these interactions on the adsorption dynamics and on alterations of the layered substrate caused by the adsorption was followed as variations of the optical film thickness (RIfS), mass increase and viscoelastic behavior (QCM-D), and electrosurface properties (MES). Experimental Section Experimental Techniques. The microslit electrokinetic setup (MES) takes advantage of one of the electrokinetic phenomena. A tangential mechanical flow of electrolyte solution at a planar solid surface leads to a distortion of the interfacial distribution of charge carriers, which is measured as streaming potential or streaming current.19,20 As shown in Figure 1, a narrow rectangular channel was formed by two sample films prepared on top of carriers (20 × 10 mm2), which was precisely aligned to be parallel, sealed, and connected to the liquid reservoir system. A Poiseuille flow was produced in dependence on a given pressure drop across the channel and induces a streaming potential Us (for a nonconductive circuit) or a streaming current Is (for a conductive circuit). The electrical potential of the hydrodynamic phase boundary ζ is derived from the Poisson equation using Us or Is according to
ζ(Us) )
(
)
η dUs 2Kσ KB + vr d p h
ζ(Is) ) -
η dIs L vr d p A
(1)
(2)
KB and Kσ represent the conductivity of the bulk solution and the surface conductivity, respectively. The channel geometry is expressed as the height h, the length L, and the cross-sectional area A () b‚h). η, v, and r are the viscosity, the permittivity of vacuum, and that of the solution, respectively. The slopes of Us and Is against the pressure drop were employed for the evaluation to eliminate electrode polarization effects. The surface conductivity was determined with changing the channel height by using the following relation from eqs 1 and 2:
nd )
miλi 2
(4)
where nd is the optical thickness of the layer (n is the refractive index and d is the physical thickness), λi is the extremum on the spectrum, and mi is the order number of the ith extremum defined as follows:
mi )
λi+1 2(λi+1 - λi)
(5)
The thickness variation of the layer is, therefore, followed as an alteration of an extremum wavelength. The reflected spectra were collected via the same fiber. For the analyses, the wavelength range between 400 and 800 nm was used in this study. (See also refs 24-27.) Sample Carriers. For the MES extended with the in situ RIfS measurements, sample polymer thin films were prepared on glass carriers (20 × 10 mm2, Berliner Glas KGaA Herbert Kubatz GmbH & Co., Berlin, Germany) precoated with a 40 nm Ta2O5 layer and a 450 nm SiO2 layer. The SiO2 layer was applied for compensating the thin thickness of the polymer films to obtain interference patterns in the visible wavelength range (see above). The Ta2O5 layer of the refractive index 2.1 was inserted between the glass and the SiO2 in order to achieve a strong reflectance at the SiO2/Ta2O5 interface. For QCM-D, 5 MHz sensor quartz crystals (Q-Sense AB, Gothenburg, Sweden) were used as sample carriers, coated by a SiO2 layer of 50 nm in thickness (GeSiM GmbH, Dresden, Germany). Polymer Films. Thin films of poly(octadecene-alt-maleic acid) (POMA, Polysciences Inc., Warrington, PA) and poly(propene-alt-maleic acid) (PPMA, Leuna-Werke AG, Germany) were prepared on top of the sample carriers as follows: anhydride POMA and PPMA were dissolved in tetrahydrofuran (0.08 wt %) and methylethyl ketone (0.1 wt %), respectively, spin-coated on the carriers, which had been modified with 3-aminopropyldimethylethoxysilane (APDMES, ABCR GmbH & Co. KG, Karlsruhe, Germany), and annealed at 120 °C for 2 h to form stable imide bonds to the modified glass surface. Remaining anhydride moieties were subsequently hydrolyzed by autoclaving prior to the adsorption experiments. Physicochemical properties of the films were reported elsewhere.28,29 In this study, the density of covalent attachments to the
Hydrophobic and Electrostatic Interactions
J. Phys. Chem. B, Vol. 110, No. 24, 2006 12121
APDMES layer was adjusted in order to obtain similar isoelectric points for POMA and PPMA films. Protein Adsorption and Desorption. Adsorption experiments of fibronectin (FN) were performed with POMA and PPMA films. A pair of the polymer samples was installed at the MES, where the channel height h (surface distance) was adjusted at 500 µm in order to keep a sufficient amount of solution and FN. After reaching a steady state in phosphatebuffered saline (PBS, pH 7.4, Sigma), 50-200 µg/mL of FN were dispensed at the polymer substrate and kept for 16 h for adsorption. Desorption was performed for 3 h, terminating the adsorption by rinsing with 20 mL of PBS. In situ adsorption/ desorption kinetics were observed as optical thickness variations using RIfS. The adsorbed mass (Γ) was estimated from the obtained optical thickness increase using de Feijter’s formula
Γ)
∆ nprotein - nPBS nads dn/dc
(6)
where ∆ is the measured increase of the optical layer thickness, n is the refractive index, and dn/dc is the refractive index increment. In this study the refractive indexes of 1.53 for the protein, 1.33 for the PBS, and 1.36 for the adsorbed layer were applied. Also dn/dc ) 1.80 cm3/g was used according to previous studies. For details, see ref 30. Surface conductivity Kσ values and isoelectric points (IEPs) were determined for both bare polymer substrates and FN-coated surfaces by the MES. The Kσ values were obtained by changing the channel height (h ) 15-50 µm) at pH 6 and 9 with 10-4 M KCl background solution, and the IEPs were evaluated from pH titration characteristics of the potential ζ in 10-2 M KCl solution (h ) 50 µm). A series of experiments was done without dewetting the sample surfaces. All experiments were performed at room temperature. After a series of measurements, the samples were investigated by confocal laser scanning microscopy (staining with fluorescent-labeled antibody TRITC), and the homogeneous distribution of FN was confirmed. FN adsorption at the copolymer films was also performed with QCM-D (Q-Sense D300, Q-Sense AB, Gothenburg, Sweden) to obtain the dynamics by determining both frequency and dissipation changes. After baseline equilibration with PBS, the polymer surfaces were exposed to 50 µg/mL of FN solution for 16-20 h for adsorption. Desorption behavior was confirmed by rinsing with sufficient amounts of PBS. Temperature was controlled within (0.01 K at the measurements (at room temperature). Fibronectin solution was purified from adult human plasma following the protocol of Brew et al.31 Results and Discussion Characteristics of Maleic Acid Copolymer Films. Alternating maleic acid copolymers with octadecene (POMA) or propene (PPMA) are composed of alternatively aligned units of maleic acid and an alkyl containing second repeating unit. The employed comonomers change the hydrophobicity of the copolymer as a whole through the variation of size and shape of the alkyl side chains (see also our previous studies29). The copolymer films showed advancing water contact angles of 100 ( 5 ° for POMA and 38 ( 5 ° for PPMA, whereas the isoelectric points (IEPs) exhibited similar values of pH 1.9 ( 0.1 (Table 1) for the utilized copolymer films (note that the IEP can be influenced by the choice of aminosilane precoating and the conditions of the copolymer film preparation): although both films contain acidic groups, the surfaces showed distinct
TABLE 1: Surface Conductivity in 10-4 M KCl (Kσ) and Isoelectric Points (IEP) of Copolymer Thin Films with and without Adsorbed FN Layers POMA Kσ at pH 6
Kσ at pH 9
polymer surface 3.4 nS 6.3 nS with 50 µg/mL FN 4.4 10.7 with 100 µg/mL FN 4.7 8.4 with 200 µg/mL FN
PPMA IEP 1.9 4.2 4.2
Kσ at pH 6
Kσ at pH 9
IEP
21.2 nS 7.1 4.8 6.7
59.6 nS 17.4 23.0 24.0
1.9 2.9 3.2 3.2
differences in their characteristics depending on the hydrophobicity of the comonomers. The large surface conductivity Kσ of PPMA films indicated a large number and/or mobility of ions at the interface, that is, a drastic swelling or shrinking of the films in response to changes of the pH and ionic strength of the solution. However, POMA was found to be a confined film. The Kσ data were found to be nearly an order of magnitude lower than those of PPMA even though the dry-state thickness was rather similar for both films (around 3-5 nm). This obviously has to be attributed to a force balance within the immobilized polymer films. The magnitude of electrostatic repulsions within PPMA films depends on the degree of dissociation of the acid groups and substantially influences the structure (extension) of the polymer film while strong hydrophobic interactions do not allow for comparable changes in POMA films due to electrostatic forces.28 The above-mentioned characteristics of PPMA and POMA films can be considered to determine the adsorption and interfacial properties of proteins as well. Fibronectin Adsorption at Hydrophobic POMA Films. Adsorption dynamics of fibronectin (FN) at the POMA film are demonstrated in Figure 2a and b referring to in situ RIfS and QCM-D measurements. Alterations of the electrosurface properties caused by the adsorption are shown in Figure 2c and Table 1. Both the optical thickness variation and the frequency decrease (i.e., mass increase) consistently indicated large adsorbed amounts of FN and hardly any desorption, which points at a strong interaction between the protein and the surface. The surface coverage of FN reached saturation at FN solution concentrations of less than 50 µg/mL on POMA (see Figure 2a and c). The IEP of the protein-covered surface approached pH 4.2 giving a sharp contrast to the acidic POMA (IEP pH 1.9). According to the optical layer thickness and the adsorbed mass,32 FN completely covers the POMA surface and, thus, solely determines the magnitude of ζ and the IEP. Namely, the obtained pH dependence of ζ and the IEP reflect the amphoteric charge characteristics of adsorbed FN. As the IEP of the FN layer approaches the intrinsic IEP of FN (∼pH 5),33 we conclude on an arbitrary orientation of the adsorbed FN at the POMA film and hardly any impact of electrostatic interactions on the structure of the adsorbed protein layer. These characteristics may be attributed to the predominance of the hydrophobic attraction between the octadecene groups of POMA and hydrophobic moieties of FN. In line with this, the high affinity between FN and POMA was also obvious by the fast saturation of the surface with protein (within 2 h). However, rearrangements of the adsorbed FN, to increase the exposure of hydrophobic protein domains to the polymer surface, may follow the initial attachment process during a longer time scale. A certain degree of desorption was found as a decrease of the optical thickness after shorter adsorption periods while the desorbed amounts were found to be reduced with increasing adsorption time (data not shown); minor conformational changes may contribute to the tight binding between the adsorbed FN and POMA. Also, both
12122 J. Phys. Chem. B, Vol. 110, No. 24, 2006
Osaki et al.
Figure 2. Characterization of FN adsorption on POMA (a, b, c) and PPMA (d, e, f). (Top): The adsorption kinetics was followed by RIfS at FN solution concentrations varying from 50 to 200 µg/mL; desorption was performed by rinsing with pure PBS (at the arrows). Measured optical thicknesses of FN layers were converted in adsorbed mass using de Feijter’s equation30 (see the Methods section). The results correlate well with our own previous amino acid analysis data.1 (Middle): The adsorption kinetics was also followed in mass (frequency) and viscoelastic properties (dissipation) by QCM-D at FN solution concentration of 50 µg/mL. The arrows indicate the desorption process. (Bottom): Acid-base characteristics (pH dependence of ζ) were compared prior to (cross) and after the FN adsorption onto POMA and PPMA at FN solution concentrations of 50 µg/mL (circle), 100 µg/mL (triangle), and 200 µg/mL (square).
the low magnitude of the surface conductivity and the small increase of the dissipation indicate that the FN layer on top of POMA films is rather confined, which agrees well with the strong hydrophobic binding of FN. Fibronectin Adsorption at Polar PPMA Films. The adsorption dynamics of FN at the PPMA film are shown in Figure 2d and e. Lower amounts of adsorption and significant desorption were observed in that case, which defines a clear contrast to POMA. The results reflect weaker attraction between the protein and the surface compared to the more hydrophobic POMA. Electrostatic interactions determine the FN adsorption at this surface leading to an adsorption of the negatively (net) charged FN against a like charge of the PPMA surface (at pH 7.4). Thus, the protein may only adsorb through preferential orientation of positively charged domains toward the polymer. Correspondingly, the IEP of the FN covered surface shows a
rather acidic value of about pH 3 (see Table 1), which may indicate the arrangement of FN molecules according to electrostatic interactions.34 The rather low affinity between FN and PPMA was supported by the slow progress of the adsorption. The saturation of the surface was not achieved yet even after 16 h (as demonstrated with both RIfS and QCM-D data), and the adsorbed amount was further increased with increasing the FN concentration up to 200 µg/mL (Figure 2d). The surface conductivity of the layered substrate underwent a remarkable decrease during the adsorption of FN onto PPMA (Table 1). As mentioned above, electrostatic repulsion between maleic acid groups stimulates the PPMA layer to expand, resulting in a high mobility of counterions reflected by high Kσ values. (Note that the FN adsorption was performed in PBS where the PPMA film attends a rather collapsed state due to the reduced Debye length; therefore, the electrostatically driven
Hydrophobic and Electrostatic Interactions
Figure 3. Variations of the viscoelastic properties as a function of the frequency shift during FN adsorption onto POMA and PPMA (from QCM-D data in Figure 2b and 2e). The plots are considered to represent the specific footprint of viscoelastic behavior on an individual adsorption process.
expansion is diminished.35 The low Kσ values of this FN-covered layer observed at lower ionic strength of the solution, however, indicates that the adsorbed protein interferes with the expansion of the PPMA film, probably due to the reduction of electrostatic repulsion within the polymer layer.) Interfacial Modes of Adsorbed FN at POMA and PPMA. The reported results confirm that the patterns of FN adsorption are significantly influenced by the dominating type of interaction between the protein and the solid surface. The adsorption dynamics on the hydrophobic POMA approached a plateau within a few hours followed by a minor conformational relaxation, whereas on PPMA the adsorption continued over a longer time period due to the rather weak electrostatic net attraction (Figure 2, RIfS and QCM-D data). Also, distinct differences were observed in the adsorbed amounts for a given FN solution concentration. The saturation of the surface with FN (monolayer-coverage, indicated by an invariant IEP) was achieved at 50 µg/mL for POMA and 100 µg/mL for PPMA. These findings demonstrate the higher affinity and binding strength of FN at hydrophobic surfaces compared to the weak electrostatic (and only partially attractive) interactions in the investigated system. Importantly, the interaction dominating the adsorption of the protein determines not only the adsorbed amount but also its “interfacial mode”, that is, the interfacial state of the adsorbed proteins at the respective surfaces. This was obvious in our study from the electrosurface properties of the compared protein covered substrates (Figure 2c and f), in particular from the shift of the IEP: more acidic characteristics of the adsorbed FN layers were obtained on PPMA even though a comparable monolayer coverage was present on PPMA and POMA (confirmed by RIfS data and a previous amino acid analysis11). Plotting the dissipation versus frequency change derived from QCM-D experiments (Figure 3) likewise revealed characteristic differences in the viscoelastic properties of the FN layers on PPMA and POMA. From these data, we conclude that FN adsorbed on the hydrophobic POMA surface loses a considerable fraction of its hydrating water shell, which is in line with the rather low surface conductivity and dissipation change of this layer. The conservation of the acid-base properties of FN on POMA, however, points at the absence of a preferential orientation of the adsorbed protein and may furthermore indicate the absence of dramatic structural changes of the protein in the adsorbed
J. Phys. Chem. B, Vol. 110, No. 24, 2006 12123 state. In contrast, FN is considered to unfold during adsorption on the polar PPMA substrate induced by electrostatic interactions6 resulting in a rather acidic FN-coated surface. Furthermore, FN appears to maintain its hydration when adsorbed onto PPMA because strong hydrophobic interactions are absent in this case: the dissipation value of the FN layer was almost threefold larger on PPMA than that on POMA, indicating a more flexible/soft adsorbed protein layer. Although differences in the binding strength of FN to POMA and PPMA have been shown already in previous studies, electrosurface characterization combined with RIfS and QCM-D provided new insight into the balance of interactions involved in the adsorption onto the compared substrates causing different interfacial modes of the protein. Because conformation, orientation, and anchorage of proteins strongly influence their functionality, the results discussed above may be important to comprehending the varying functionality of the protein when adsorbed to different surfaces. Conclusions Fibronectin (FN) adsorption/desorption was investigated at hydrophobic POMA and polar PPMA surfaces. It was demonstrated that the interfacial mode of the adsorbed FN significantly depends on the predominant attractive forces between the protein and the surface. Hydrophobic interaction directs the adsorption on POMA, resulting in higher adsorption rates as well as larger adsorbed amounts and less desorption compared to PPMA where electrostatic forces dominate the protein-substrate interactions. On PPMA, FN was found to adsorb against a similar net charge of protein and substrate pointing at the fact that electrostatic matching of positively charged amino acid side chains with maleic acid groups has to be involved. Differences of the binding strength of FN onto POMA and PPMA reported in earlier studies were clarified in this work by dedicated electrokinetic experiments revealing the dominating interactions between the protein and each of the compared polymer substrates. The results underline that careful consideration of physicochemical surface characteristics of materials may provide important clues on the functionality of adsorbed protein layers as these features “translate” properties of the underlying substrate toward cells adhering on top of the adsorbed protein layers. References and Notes (1) Hynes, R. Fibronectins; Springer-Verlag: New York, 1990. (2) Ho¨rmann, H. Fibronectin - Mediator between Cells and ConnectiVe Tissue; Klinische Wochenschrift: 1982; Vol. 60, pp 1265-1277. (3) Magnusson, M. K.; Mosher, D. F. Arterioscler., Thromb., Vasc. Biol. 1998, 18, 1363. (4) Meadows, P. Y.; Walker, G. C. Langmuir 2005, 21, 4096. (5) Faucheux, N.; Schweiss, R.; Lu¨tzow, K.; Werner, C.; Groth, T. Biomaterials 2004, 25, 2721. (6) Bergkvist, M.; Carlsson, J.; Oscarsson, S. J. Biomed. Mater. Res. 2003, 64A, 349. (7) Keselowsky, B. G.; Collard, D. M.; Garcia, A. J. J. Biomed. Mater. Res. 2003, 66A, 247. (8) Michael, K. E.; Keselowsky, B. G.; Vernekar, V. N.; Meredith, C. A.; Latour, R. A.; Garcia, A. J. Langmuir 2003, 19, 8033. (9) Katz, Z.; Zamir, E.; Bershadsky, A.; Kam, Z.; Yamada, K. M.; Geiger, B. Mol. Biol. Cell. 2000, 11, 1047. (10) Garcia, A. J.; Vega, D. V.; Boettiger, D. Mol. Biol. Cell. 1999, 10, 785. (11) Renner, L.; Pompe, T.; Salchert, K.; Werner, C. Langmuir 2005, 21, 4571. (12) Renner, L.; Pompe, T.; Salchert, K.; Werner, C. Langmuir 2004, 20, 2928. (13) Pompe, T.; Markowski, M.; Werner, C. Tissue Eng. 2004, 10, 841. (14) Pompe, T.; Renner, L.; Werner, C. Biophys. J. 2005, 88, 527.
12124 J. Phys. Chem. B, Vol. 110, No. 24, 2006 (15) Pompe, T.; Kobe, F.; Salchert, K.; Jørgensen, B.; Oswald, J.; Werner, C. J. Biomed. Mater. Res. 2003, 67A, 647. (16) Zimmermann, R.; Osaki, T.; Kratzmu¨ller, T.; Gauglitz, G.; Dukhin, S. S.; Werner, C. J. Am. Chem. Soc., submitted for publication. (17) Zimmermann, R. Charakterisierung Von Ladungsbildungsprozessen an Polymeren in wa¨ssrigen Lo¨sungen. Ph.D. Thesis, Technische Universita¨t Dresden, Germany, 2004. (18) Zimmermann, R.; Osaki, T.; Gauglitz, G.; Werner, C. Proceedings of the 7th Dresdener Sensor-Symposium; TUD press: Dresden, 2005. (19) Lyklema, J. Fundamentals of Interface and Colloid Science; Academic Press: London, 1995; Vol. 2. (20) Hunter, R. Zeta Potential in Colloid Science; Academic Press: London, 1981. (21) Werner, C.; Ko¨rber, H.; Zimmermann, R.; Dukhin, S. S.; Jacobasch, H. J. J. Colloid Interface Sci. 1998, 208, 329. (22) Zimmermann, R.; Dukhin, S. S.; Werner, C. J. Phys. Chem. B 2001, 105, 8544. (23) Werner, C.; Zimmermann, R.; Kratzmu¨ller, T. Colloids Surf., A 2001, 192, 205. (24) Gauglitz, G. Biosensors for EnVironmental Monitoring; Bilitewski, U., Turner, A. P. F., Eds.; Harwood Academic Publishers: Amsterdam, 2000; pp 28-51.
Osaki et al. (25) Birkert, C.; Tu¨nnemann, R.; Jung, G.; Gauglitz, G. Anal. Chem. 2000, 74, 834. (26) Haake, H. M.; Schu¨tz, A.; Gauglitz, G. Fresenius’ J. Anal. Chem. 2000, 366, 576. (27) Schmitt, H. M.; Brecht, A.; Piehler, J.; Gauglitz, G. Biosens. Bioelectron. 1997, 12, 809. (28) Osaki, T.; Werner, C. Langmuir 2003, 19, 5787. (29) Pompe, T.; Zschoche, S.; Salchert, K.; Herold, N.; Gouzy, M. F.; Sperling, C.; Werner, C. Biomacromolecules 2003, 4, 1072. (30) Vo¨ro¨s, J. Biophys. J. 2004, 87, 553. (31) Brew, S. A.; Ingham, K. C. J. Tissue Cult. Methods 1994, 16, 197. (32) A surface coverage of approximately 250 ng/cm2 is determined by an assumption of densely packed monolayer of FN dimers, with dimensions of 60 nm in length, 2.5 nm in diameter, and 220 kDa for the molecular weight as a monomer. (33) Tooney, N. M.; Mosesson, M. W.; Amrani, D. L.; Hainfeld, J. F.; Wall, J. S. J. Cell Biol. 1983, 97, 1686. (34) Wilson, K.; Stuart, S. J.; Garcia, A. J.; Latour, R. A. J. Biomed. Mater. Res. 2004, 69A, 686. (35) Pompe, T.; Renner, L.; Grimmer, M.; Herold, N.; Werner, C. Macromol. Biosci. 2005, 5, 890.
